The Biopsychosocial Model

In brief

George L. Engel (1913 to 1999), a psychiatrist and internist at the University of Rochester, published "The Need for a New Medical Model: A Challenge for Biomedicine" in Science in 1977. He argued that the dominant biomedical model, which explains illness as a deviation in measurable biological variables and nothing else, could not account for why two patients with an identical lesion fared so differently. His replacement holds that biological, psychological, and social factors interact at every level of illness, and that leaving any one out produces bad medicine. The diathesis-stress model, developed in the same period for psychiatric disorder specifically, supplies the mechanism: a vulnerability becomes an actual disorder only once enough environmental stress is layered on top of it.

The full treatment

The problem it answers

By the mid twentieth century the biomedical model had produced antibiotics, anesthesia, and a run of dramatic cures, and its method, find the physical lesion or biochemical defect and correct it, was medicine's working creed. Engel did not dispute the successes. He argued the model's exclusivity was the problem: by design it treated the psychological and social dimensions of illness as irrelevant to diagnosis and treatment. Two patients with the same degree of coronary artery disease can differ enormously in disability and survival, differences that track depression and social isolation as much as plaque. Psychiatry, leaning hard by the 1970s on new drugs like chlorpromazine, risked treating a diagnosis as a pure brain lesion and losing the person attached to it. Engel wanted a model wide enough to hold both.

How it works

Engel borrowed a systems hierarchy from the biologist Ludwig von Bertalanffy (1901 to 1972), whose general systems theory described nature as nested levels, atom, molecule, cell, organ, organism, each obeying its own laws while open to influence from the levels above and below. Engel extended the hierarchy past the organism to the person, the family, the community, and the culture. A change at any level, a mutation, a bereavement, a job loss, a neighborhood's poverty, propagates both up and down the chain. Illness is never purely biological or purely social; it is a transaction across levels, and treating a patient well means attending to the biological, the psychological, and the social together, asking how each shapes the others in this particular case.

What it claims

The model's most testable descendant is the diathesis-stress model, built to explain why only some biologically vulnerable people become ill. A diathesis is a predisposition, genetic, neurodevelopmental, or psychological, that produces no disorder by itself. Paul Meehl (1920 to 2003) proposed in 1962 that an inherited neural defect he called schizotaxia produces, under ordinary rearing, a personality organization he called schizotypy, which decompensates into clinical schizophrenia only under added stress. Joseph Zubin (1900 to 1990) and Bonnie Spring generalized this in 1977 into a vulnerability model: everyone carries some threshold, and an episode occurs once life stress exceeds it. The logic is now standard across psychiatry, applied to depression, PTSD, and substance use, which is why an identical stressor, a divorce, a deployment, produces a lasting disorder in one person and nothing durable in another.

The key study or demonstration

The clearest empirical test of the diathesis-stress logic came from Avshalom Caspi, Terrie Moffitt, and colleagues, who reported in Science in 2003 that a functional variant in the promoter region of the serotonin transporter gene (5-HTTLPR) moderated the effect of stressful life events on depression, in a New Zealand cohort followed from birth into their mid twenties. Carriers of the short allele showed more depressive symptoms after stressful events than those with two long alleles; among people with few stressors, genotype barely mattered. Neither the gene nor the stress alone predicted depression well, only the interaction did, exactly the diathesis-stress claim: vulnerability sits latent until stress activates it. On the biopsychosocial side, Engel gave his own demonstration in the 1980 paper, walking through a man recovering from a heart attack whose subsequent cardiac arrest, Engel argued, was precipitated not by a new biological event alone but by the anxiety of an intrusive blood draw at a moment of helplessness, a psychological trigger acting on a compromised biological substrate.

Related distinctions

The diathesis-stress model is often mistaken for a simple additive risk model, but its claim is interactive, a gene by environment interaction: vulnerability and stress multiply each other's effect rather than adding up. A later refinement, differential susceptibility theory, proposed by Jay Belsky and elaborated with Michael Pluess in 2009, argues that the same trait making some people more harmed by adversity also makes them more helped by support, so-called orchid children who wilt in bad environments and bloom in good ones, against hardier dandelion children who fare adequately regardless. This reframes vulnerability as heightened sensitivity to environment in general, a susceptibility that cuts both ways.

Lineage

Engel's model descends from two traditions. Its systems side comes from Bertalanffy's general systems theory (1968), which supplied the hierarchy of nested levels. Its clinical side comes from Adolf Meyer (1866 to 1950), the Swiss-born psychiatrist who chaired the Phipps clinic at Johns Hopkins and argued decades earlier for a "psychobiology" treating mental disorder as a whole person's reaction to a life history, not a discrete disease entity. Engel's mentor was John Romano (1908 to 1994), who trained under Meyer and recruited Engel to found Rochester's division of psychosomatic medicine in 1946, where the model was worked out over three decades before publication. The diathesis-stress lineage runs separately through the older clinical observation, back to Emil Kraepelin and Eugen Bleuler, that predisposition and precipitating stress jointly explain the onset of psychosis, an observation Meehl and then Zubin and Spring turned into a formal model.

The strongest case for it

The model's durability rests on a plain clinical fact the biomedical model does not explain: identical diagnoses produce wildly different courses, and the difference is rarely biology alone. It gave medicine a vocabulary for the placebo effect, for why depression predicts worse cardiac outcomes independent of disease severity, and for why attending to a patient's family, housing, or income changes clinical results. Collaborative care programs that pair medication for depression with structured psychological support, built on this logic, show measurable gains over medication alone in controlled trials. The model also legitimized entire fields, health psychology, psychosomatic medicine, consultation-liaison psychiatry, for which biomedicine alone had no room.

The strongest case against it

The sharpest critique comes from S. Nassir Ghaemi, a psychiatrist at Tufts, in The Rise and Fall of the Biopsychosocial Model (2009). Ghaemi argues that Engel's model degenerated into what he calls "biopsychosocial-ism": an instruction to weigh biology, psychology, and society in every case, with no principle for weighing them against each other. A framework that tells a clinician everything matters somewhat, he argues, is not a theory that can be tested and found wrong; it is a permission slip that lets any clinician justify whatever mix of drugs, therapy, or social intervention they already preferred. He contrasts this with Karl Jaspers's older methodological pluralism, which specified which method, causal explanation or empathic understanding, applied to which kind of case, rather than blending everything indiscriminately. The diathesis-stress model has faced its own reckoning: the Caspi 2003 finding, celebrated at first as confirmation of the interactive logic, failed to replicate in a meta-analysis led by Neil Risch, published in JAMA in 2009, which found no significant interaction between 5-HTTLPR, stress, and depression risk across the pooled studies. The general logic survived that better than the specific gene did, but the episode was a lesson in how far a compelling story can outrun the evidence for its exact mechanism.

Where it stands now

The biopsychosocial model remains the official framework of American medical education, embedded in the DSM-5's cultural formulation interview and the World Health Organization's disability framework. It is rarely defended now as a precise scientific theory, and more often used, much as Ghaemi anticipated, as a checklist against reducing a patient to a lab value. The diathesis-stress model has aged better as science: it remains the standard frame in clinical psychology for schizophrenia, depression, and PTSD, refined by differential susceptibility research and by genome-wide approaches that replaced single candidate genes like 5-HTTLPR with polygenic risk scores measured against documented adversity. The core claim both share, that a cause is never purely internal or purely external, is close to consensus in psychiatry, even where the frameworks used to state it remain contested.

Test yourself

Think of an illness or a hard stretch you have lived through, physical or psychiatric. List what a purely biological account would credit, a virus, a genetic risk, a hormone, then list what changed in your mind and your circumstances at the same time. If the second list is doing real explanatory work, you already think in biopsychosocial terms. The harder question: how would you weigh the three factors differently, and on what basis other than what felt right at the time?